We learn from experience to make more effective decisions, often adjusting our expectations to match outcomes that we have experienced in the past. In a dynamic world, this adjustment process must itself be adaptive, because sometimes changes occur that render past outcomes irrelevant to future expectations. For example, a forager must recognize that a dying tree will no longer yield as much fruit as when it was healthy. The goal of the proposed research is to test the novel hypothesis that certain individual differences in decision- making, including some that depend systematically on age, result from differences in how observed changes in the world affect the use of past experiences to inform future decisions. We propose that individuals differ substantially in how they recognize and respond to two different forms of uncertainty. One form of uncertainty, called noise, represents random fluctuations in an otherwise stable process. The other form of uncertainty, called volatility, represents fundamental changes in the process itself. Individual, adaptive decision-making ranges from a tendency to always adjust expectations in the face of either noisy or volatile new data, to a tendency to form stable expectations that are relatively unaffected by new data. We have identified computational principles that govern this kind of adaptive belief updating, which include the importance of prior expectations about the rate of occurrence of volatile changes in the current environment. We also propose that the important underlying computations are encoded by the activity of neurons in the anterior cingulate cortex (ACC). To test these hypotheses, we use two complementary Aims that provide multiple measures of ACC activity in individual subjects of different ages performing tasks designed to measure directly the amount of influence each new piece of information has on existing beliefs about a dynamic environment. The first Aim uses combined behavior and neurophysiological recordings in younger and older monkeys, providing measurements of the relevant neural computations with high spatial and temporal resolution. The second Aim uses combined behavior and fMRI- and EEG-based measurements of brain activity in human subjects, which can be compared directly to the higher-resolution monkey data. The human studies also include several versions of the task, including one designed to manipulate and measure prior expectations about the rate of environmental changes directly, that provide a broader view of the individual and age-related differences that occur with these kinds of decisions. The proposed research represents a novel field of study that is likely to provide far-reaching insights into how individuals make effective decisions in a dynamic world.